1887

Abstract

Summary

In this work we present the results of a preliminary safety assessment of an artificial intelligence assistant designed to support drilling and well engineers to write detailed operational procedures (DOP). We assessed the prototype of the assistant through empirical testing of the system in context with 5 end users, through interviews with subject matter experts, and by exploring previously reported incidents related with DOPs in existing databases. The findings shown that the AI assistant would introduce new risks to the system related with differents types of errors in the outputs. Even though other pre-existing barriers (linked to work processes and human supervision) could prevent a serious incident, it is deemed that this technology is not mature for use in high risk industries and safety critical tasks. We discuss the findings and present recommendations for future development and assessment.

© EAGE Publications BV
Loading

Article metrics loading...

/content/papers/10.3997/2214-4609.202639050
2026-03-09
2026-02-15

  • From This Site

    /content/papers/10.3997/2214-4609.202639050
    dcterms_title,dcterms_subject,pub_keyword
    -contentType:Journal -contentType:Contributor -contentType:Concept -contentType:Institution
    10
    5
Loading full text...

Full text loading...

References

  1. Bainbridge, L. (1983). Ironies of automation. In Analysis, design and evaluation of man-machine systems (pp. 129–135). Pergamon.
     [Google Scholar]
  2. Endsley, M. R. (2023). Ironies of artificial intelligence. Ergonomics, 66(11), 1656–1668.
     [Google Scholar]
  3. Gergerechi, A., SørensenE., & Jensen, J.E. (2017). Report of the investigation of the well control incident in well 31/2-G-4 BY1H/BY2H on the Troll field with the Songa Endurance drilling unit. https://www.havtil.no/contentassets/dcc7039f921c41a8af25794c3e8f9a35/investigation-report-statoil-troll_songa-endurance, last retrieved on 01.07.2025
     [Google Scholar]
  4. Kirwan, B. (2024, November 26th). Supporting Human-AI Teaming: What Questions Should We Be Asking? Retrieved from https://skybrary.aero/webinars/supporting-human-ai-teaming-what-questions-should-we-be-asking
     [Google Scholar]
  5. Naveed, H., Khan, A. U., Qiu, S., Saqib, M., Anwar, S., Usman, M.,… & Mian, A. (2023). A comprehensive overview of large language models. ACM Transactions on Intelligent Systems and Technology.
     [Google Scholar]
  6. OpenAI (2023). GPT-4V(ision) System Card. https://cdn.openai.com/papers/GPTV System Card.pdf, last retrieved on 07.07.2025
     [Google Scholar]
/content/papers/10.3997/2214-4609.202639050
Loading
Safety Insights on Generative AI Assistants: A Case Study in Drilling and well Intervention
Conference Proceedings 2026, 1 (2026); https://doi.org/10.3997/2214-4609.202639050
/content/papers/10.3997/2214-4609.202639050
/content/papers/10.3997/2214-4609.202639050
Loading

Data & Media loading...

Most Cited This Month Most Cited RSS feed

This is a required field
Please enter a valid email address
Approval was a Success
Invalid data
An Error Occurred
Approval was partially successful, following selected items could not be processed due to error